Fixing site problems found by SEO by Zaweba

When SEO by Zaweba crawls your own site, it doesn't just score pages one at a time. It looks for problems that only show up across the whole site: orphan pages, pages linked but missing from the sitemap, duplicate titles and descriptions, near-duplicate pages, thin pages, noindexed pages listed in the sitemap, canonical conflicts, redirect chains and pages returning 4xx or 5xx errors. It also suggests links to add.

These are the housekeeping problems that build up quietly on any site that's been around for a while. Each is small on its own. Together they can make a site harder to crawl and understand.

Getting the crawl to run

The site crawl is part of Step 4, alongside the competitor crawl, and it's optional. It reads robots.txt, reads your XML sitemaps (including sitemap indexes and gzipped ones) and crawls breadth-first up to the page limit, which is up to 500 pages by default and can be changed in Settings. It builds the internal link graph as it goes, with inbound links, click depth and anchors for each page.

The problems list comes from that crawl, so the more of your site it reaches, the more complete the list is.

Problems with links and sitemaps

An orphan page has no internal links pointing to it. Visitors can't find it by clicking around, and search engines may struggle too. Pages linked but not in the sitemap are the opposite: reachable by links but missing from the list you give search engines. Noindexed pages in the sitemap send mixed signals, asking for a page to be crawled while telling search engines not to index it.

For these, the link suggestions help. Where a page needs more internal links, the app suggests where they could come from, so you're not guessing which pages should point to which.

Problems with duplicates and thin pages

Duplicate titles and descriptions make pages hard to tell apart in search results. Near-duplicates are found with SimHash, a way of spotting pages whose text is very similar even when it isn't identical. Thin pages have very little content. Canonical conflicts happen when pages point their canonical tags in ways that contradict each other or the rest of the site.

These usually need a decision rather than a quick fix. Should two near-duplicate pages be merged, or made genuinely different? Should a thin page be expanded, or folded into a stronger one?

Redirects and broken pages

Redirect chains are links that hop through several redirects before landing. They slow things down and waste crawling. Pages returning 4xx or 5xx errors are broken or failing. Both are usually quick to fix once you know where they are: point links straight at the final address, and fix or remove broken pages.

Working through the list

Each problem becomes a check with a severity, an effort and a plain reason, and feeds the action plan. The persistent to-do list lets you Mark done, Defer, Dismiss with a reason, or Reopen each task. At the next scan, done tasks are checked again and reopened if the check still fails, so nothing quietly slips back. Export the table with File, then Export table as CSV (Ctrl+E), if you'd like to work through the problems in a spreadsheet or hand them to a developer.